Architecture Made Easy, Part 4

As the current bleak economic climate reminds us, the business marketplace is a harsh, Darwinian struggle to survive. Companies that are “household names” in boom times are swept away
the moment they fail to bend to the winds of economic change.

But for all the talk of an ultra-competitive marketplace, there is an awful lot of commonality to be found among competitors. After all, we live in a world where management is educated in a handful
of elite universities and base their decisions upon identical advice from the same high-priced advisors – who themselves were educated in the same elite universities.

It should therefore come as no surprise that the competitive advantages that spell life or death for an enterprise are often both vanishingly small (large innovations are rapidly imitated by
competitors) and almost always exist outside of the “group-think” of the current industry paradigm (if it is “best-practice” everyone else is already doing it).

For those who doubt the debilitating power of this tendency for group-think, they need only look at the results when management asks for insight into the operations of the enterprise. In response,
management receives the current business process model (BPM) along with the metrics which show how an operation is conforming to the BPM plan. Note they do not receive, nor think to ask, for any
insight into whether this is the correct plan for operations to be pursuing, only whether the BPM is being executed as planned.

Consider if a military commander asked his staff how the left flank was faring in battle, and he received only metrics demonstrating how well his troops were sticking to the battle plan. One would
hope that our fictional commander would have the presence of mind to ask, “It is great that we are following the battle plan… but how is the battle plan working?”

As the Persian King Xerxes must have realized, as he watched the grossly outnumbered Greeks annihilate his naval fleet in the battle of Salamis:

“No plan survives contact with the enemy.” (Helmuth von Moltke)

A good business process model (BPM) only gets your operation so far. The business processes of Circuit City and Best Buy may have been nearly identical, but one looks likely to survive the current
economic downturn, while the other is filing for bankruptcy.

As operational inefficiencies have already impacted Circuit City’s general ledger, it is fairly straightforward to identify what Circuit City “should have done.” Less obvious is
how management could have identified those issues in time to improve their likelihood of survival.

Although BPM can generally have a useful role, its use is not enough. Companies can certainly use business process modeling to help design their operational processes and using the latest tools,
managers can run simulations on the business operation they have designed. However, no matter how much a modeling tool may help someone design their business operation, it does not tell them how to
manage and run that business operation in response to real events. Counting on a model to help someone manage a business operation is no more helpful than a TomTom could be in revealing the safest
route to take while driving in the midst of a hurricane. The simple reality is that simulators and pre-planned models have neither use for radar, nor late-breaking news reports.

With the collapse of AIG, there has been a great deal of focus on the insurance industry. Within insurance, there are an abundance of business process models that depict the operational processes
of property and casualty, and life insurance. However, BPM will never tell you what business and economic conditions you will encounter each day, and BPM simply cannot account for the differences
that exist between each and every member of your operational team, any more than a flight simulator can help you when you are sitting in the cockpit of an aircraft with real up drafts, cross winds,
air temperatures and obstacles to avoid, while you are simultaneously dealing with a new flight crew.

Once we enter the real world, we need radar, whether that radar illustrates your true physical environment, or your live business and operational environment. Either way, there is no
substitute for objective real-time feedback of what is happening around us.

As the spirit of Edward John Smith, captain of the Titanic, would tell us today, there is no substitute for having radar, no matter how well your ship is designed.


Enter Operational Metadata

In business, the easiest way to graphically visualize business radar is with business intelligence (BI). The business value of applying business intelligence to operational processes instead of
transactional and financial data is surprisingly critical to the enterprise and has largely gone unnoticed and, consequently, untapped in the business world. Those few enterprise architects (EA)
that have already recognized that operational process data is rich in business value, will be best positioned to deliver a powerful competitive advantage to their organizations.

To explain in more detail, operational metadata encompasses all of the information surrounding, yet excluding, transactional business data. It includes information like the types of operational
requests your business operation received during each moment of the day, where they originated, who received them, how long each step took to complete and by whom, and how that compared to the same
period last year or compared to other employees.

Even the full breadth of metadata available to companies remains largely misunderstood as it is rarely identified and conspicuously absent from the marketing literature of
metadata repository products.

However, the simple fact is that operational metadata provides the sole source of business radar within an enterprise, and with a measurable ROI for business stakeholders.

Additionally, once the value of business radar on your operational processes becomes apparent, the steps that must be taken to achieve it are easy to understand and are easy to perform.


What is Metadata?

The truly boring definition of metadata has always been simply “data about data,” but the new and exciting definition comes from Forrester, an independent research firm,
defining the concept as “information that describes or provides context for data, content, business processes, services, business rules, and policies that support an organization’s
information systems.”

One reason why metadata for operational business processes is so significant is that conventional metadata simply does not deliver the expected ROI touted by major metadata repository
vendors, while the ROI on “operational metadata” is considerable and easy to determine.


A Brief History of Metadata

The earliest forms of “metadata” were collected about laborers in Egypt in 2560 BC to determine payment for artisans, and metadata was collected about documents and scrolls in 300 BC to
track the inventory of the Ancient Library of Alexandria.

The first metadata repository products appeared in the 1970s in the form of a data dictionary, which housed definitions, relationships, date of origin and data format. The role of this
metadata was to assist database administrators in managing the storage of data for automated business processes primarily from the technology perspective of database management systems
(e.g., any DBMS data dictionary, such as the ones associated with IMS, Oracle, DB2 or IDMS).

The next metadata repository products appeared in the 1980s with the development of CASE (computer-aided software engineering) tools that automated the process of designing databases and
applications. These CASE tool-based repositories were focused upon the technology perspective of design and implementation. CASE tool metadata repositories created a degree of redundancy
with the already existing data dictionary metadata repositories (e.g., CA’s ERwin Model Manager).

In the 1990s, the next generation of metadata repository products attempted to provide useful information by consolidating metadata across multiple platforms for analysis and
reporting purposes, which included large amounts of information that were derived from automated scans of application source code. These metadata repository products were used to infer
vast amounts of information from existing inventories of software and then to make it available in ad hoc analysis and reporting. Similar to CASE tool-based metadata repositories, these
products were also focused upon the technology perspective of the implementation artifacts (e.g., Platinum Repository).

In the 2000s, a new generation of metadata repository products was introduced, which were fully integrated repositories, forming a data warehouse of metadata information that encompasses
the various models and design artifacts from each step of an end-to-end, model-driven software development process. These metadata repositories were focused upon the technology perspective
of the software development life cycle (SDLC), which began by capturing business workflow and requirements within various types of business models, and then continued their collection of data by
capturing the design of applications and databases in process models, data models, and subsequent implementation models and artifacts (e.g., PowerDesigner Repository).


Metadata Repository Tools Fail

The big name metadata repository tools that scrape data definitions out of application programs, or extract information out of database data dictionaries or data models, present the user with an
interesting paradox.

By making the data redundant, they force the user to decide whether to use the data in the repository or the more current data at the source.

For example, the stock price listed in yesterday’s newspaper may still contain the correct stock price, but why take the chance?

Developers will invariably go to the source that they believe to be the most accurate and reliable, which was where the information originated back in the application program, DBMS data dictionary,
or data model in the CASE tool or CASE tool repository. And who among us can blame a developer for using the source they trust completely and thoroughly?

It is this redundancy paradox that users face that ensures that even the most well intentioned big name metadata repository cannot achieve a positive ROI.


Operational Metadata ROI is High

In contrast to the conventional metadata that is replicated within major metadata repository products, operational metadata offers a completely different value proposition to the business, as
operational metadata reveals a great deal about the health of your business and your business operation.

The confluence of business intelligence and operational metadata is uniquely able to help the business gain insight into the operational environment and business activities using graphical images
that make it easy to grasp what is occurring, not unlike the graphical image that appears on a radar scope. One of the big advantages is being able to see these graphical images well before
accounting systems feed their transactional results into your general ledger for reporting. In this manner, business radar provides an early warning system about many of the factors that influence
your bottom line that could otherwise go completely undetected.

The absence of radar is like everything flying at you below the radar.

However, operational metadata not only provides business radar within your operation, but it also provides insight externally into the marketplace, again before the resulting numbers hit your
balance sheet and subsequently appear in your financial reporting.

To put this notion in its proper perspective, all we have to do is to contemplate the time before the deployment of radar systems.

During the period predating radar, the primary method of alerting a naval commander of a possible threat was a human lookout stationed in the crow’s nest to visually spot another vessel on
the horizon. In thick fog or at night, however, the lookout might be lucky to just see the end of his arm.

In businesses predating business radar, corporations employed automation to accelerate the processes that used to be performed manually. Automation in this context was simply a way to leverage the
speed of a computer to perform routine activities to reduce the cost of each transaction that was previously performed by hand and the human mind.

It is quite common to see corporations leveraging massive amounts of computer power to automate activities within the enterprise; however, even today, there are almost no computer systems that are
assigned the critical role of providing business radar.

More significantly, let’s also consider the advanced capabilities that have been built on top of the initial radar systems.

Modern radar, sonar, seismic and sound systems have advanced considerably. Not only do today’s radar systems warn people that something is out there, but the modern radar system can also
calculate the speed and direction of the object detected, what the dimensional characteristics of each object may be, often identifying the type and model of the object or collection of objects,
including its manufacturer. Additionally, it can frequently determine its geographic origin, range and possible destination. The complete modern radar system can also determine the potential threat
level to assets in the area and automatically cut orders to optimally defend those assets. Let’s now carry this analogy over to the business world.

In the few businesses that employ business radar, they not only use computer technology to support their basic operational reporting, but they also use computer automation to illustrate the
efficiency of their business processes and the productivity of their employees. They also provide early warning mechanisms to alert management to changes that are detected in their operating
environment and can suggest adjustments in the staffing and expertise levels to address the various factors that threaten optimum profitability and performance.

When competing against a company that has business radar, the issue becomes how a business that does not have business radar survives when it cannot determine what is happening inside and
outside their business operation?

Another area of applicability for business radar is when conducting performance assessments of employees. While some of us may be busy using business process modeling to determine the current and
future design of our operational procedures and running simulations to determine the least number of steps required to underwrite an insurance policy, managers are preparing to do their mid-year
and year-end employee performance evaluations. These managers are attempting to identify who among the organization is contributing relative to others so that they may be rewarded for outstanding
contributions.

Some companies will be equipped to do objective employee performance reviews using metrics, while others will only be able to rely upon subjective criteria.

A company without business radar must rely upon rather subjective performance evaluations. Those managers will hear whatever the employees wish to share with them, including any useful perceptions
that the employee wishes to impart upon the manager. The problem is that whether these perceptions are true or false, they are not based upon objective measures.

Although they may operate without objective metrics, the odd thing here is that extremely large companies often adopt an attitude that nothing can seriously threaten them because they are so huge,
and as a result these are the very companies that are least likely to have business radar.

Contrary to this mentality, it is the extremely large enterprise that needs business radar the most, as they represent the biggest targets where a smaller more agile competitor can gain the
most advantage.

If you are employed by a huge corporation, the ability to survive in business is rarely put to the test, at least in a visible way. However, when customers become scarce, the competition gets
emboldened as they adapt to the new realities of the marketplace and they siphon the profitability out of these large companies that never seem to show the appropriate level of concern. It is for
this reason that many large companies frequently rotate in and out of the Fortune 100.

So what is the ROI, and how can it be calculated?

Unlike its earlier metadata cousins, the business ROI created by operational metadata provides significant and essential value to the enterprise that can be readily measured. For internal
operational processes, introducing workflow automation provides essential capabilities that objectively track and report the various business activities. Extremely important to many areas of
business, workflow automation also provides essential standardization of business processes. Once the workflow is automated, the ability to illustrate productivity, efficiency, error rates, and the
differences among individuals and operational units becomes a crucial tool to both senior and middle management with real ROI.

A common example of real ROI is that each human resource that is ineffectively placed within any particular operational environment may easily represent at least $150,000 annually. Upon reviewing
the operational metrics, if there are as few as four individuals not optimally placed to support the workload, then that represents over a half a million dollars of budget that could be better used
elsewhere within the organization. In an operation that consists of more than thirty staff members, it would not be surprising if four staff positions could be restructured to move toward a more
optimal business operation.

As an example, in a twenty-five member data management services department of a 4,000 employee company based in Princeton, it was found after workflow automation was put in place that nearly a
quarter of the data management services staff had not been optimally placed.

These are the types of inefficiencies and blind spots that management of any enterprise can only hope is weighing down their competition. Such inefficiencies cause their competitors prices and
their profits to be less. Moreover, competing against any enterprise that is operating without business radar is simply a much easier task.


The Big Picture

Business intelligence (BI) tools are naturally capable of helping the business user visually illustrate the metrics that present among their operational metadata. BI tools can easily illustrate the
characteristics of a business operation and its participants in objective terms real-time as a way to graphically represent the findings of business radar.

Additionally, by providing an early warning mechanism illustrating and comparing prior fiscal periods against the current period, the enterprise can begin analyzing what is happening much sooner,
before financial analysis activities are being reported in the last stages of back office automation.

There is also value in knowing where business activity originates from day to day, period to period, and what the size of the incoming business activity is at any given point in time in any fiscal
period or business cycle. Knowing the business operation’s excess capacity at every moment can also reveal when the operation is running too leanly or too richly.

Each business marketplace is in flux, just like weather and ocean currents. Being able to detect changes in the business environment, or in customer demand, can help a company better adapt to meet
the challenges of the competition.

The primary benefit of an early warning mechanism is that it affords you the opportunity to investigate and understand, whether good or bad, the root cause of your observations. It also
allows any company to operate more leanly when necessary without losing control of the operation.


Getting Started

Every company can determine for itself whether there are compelling reasons to equip their managers with business radar, so they can manage the operation more effectively and efficiently using
objective measures.

Begin by selecting business operations that have any combination of the following characteristics:

  • Are high value to the business
  • Have more than twenty staff members
  • Contain high wage earners

Experiment with BI visualization preferences that best communicate the various factors that may be important to the manager of the business operation. Like our naval analogy, try to identify the
potential causes of whatever is being illustrated, highlight its relative importance to management and even suggest a potential course of action.

To do this effectively, leverage industry expertise, and those knowledgeable in your business. Get advice from operational metadata and BI experts that can help you identify what data is available
and what kind of business radar that data can offer.


Conclusion

Business radar has arrived, and it looks surprisingly like operations management is going the way of high technology video games with visualizations and built-in advisors. It is hard to imagine
that such a normally dry topic, such as metadata, would become so interesting and valuable.

Remember, begin with the business operations, in either the business side or in IT that:

  • Are essential or of high value to the company

  • Have more than twenty staff members

  • Contain high wage earners

Think about the analogy of how military uses for radar have become so technologically advanced, such as:

  • Identify the origin of each operational activity

  • Try to assess the value of each operational activity

  • Try to assess the overall impact of each operational activity upon your available capacity

  • Investigate changes as potential shifts in the marketplace

  • Consider automatically offering recommendations to address what is showing up on your business radar

Aim high to achieve the maximum ROI that can be delivered back to the enterprise, such as:

  • Assess the value of metrics-driven management to your operation

  • Assess the reaction to accurate measures of staff member performance

  • Calculate the value of optimally positioning each staff member within the operation

Plan and evaluate your options carefully, such as:

  • Engage the most knowledgeable business expertise to facilitate the maximum benefit to the business operation

  • Attend speaking events and conferences by those that specialize in this essential capability

  • Always keep in mind that the costs should be low and the ROI high

  • Google workflow automation and business radar

Now you see what is meant when we say:

“To get business value from metadata, focus on operational processes.”

Please don’t hesitate to let me know which articles in the “Architecture Made Easy” series are useful to your organization. In addition, corrections, enhancements, and suggestions
are always welcome and are requested.

Share

submit to reddit
Top